Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    and does

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    まい
    0.73
    gister
    0.68
     foolproof
    0.67
     fathom
    0.66
     incapable
    0.66
    establish
    0.65
     위하여
    0.64
    Establishing
    0.64
     germinate
    0.64
    做到
    0.63
    POSITIVE LOGITS
     revolves
    0.93
     differs
    0.92
     acknowledges
    0.91
     reflects
    0.90
     departs
    0.90
     explains
    0.90
     remains
    0.88
     focuses
    0.86
     tries
    0.85
    ⇝
    0.84
    Act Density 0.002%

    No Known Activations